S9 Understanding mercury - organic matter interactions

Thursday, 28 July, 2011

RS9-P1 — 11:00-12:00 and 17:30-18:30
Authors: SUBIR, Mahamud1, ARIYA, Parisa A.1
(1)McGill University, 801 Sherbrooke St. West, Montreal, QC, Canada H3A 2K6, mahamud.subir@mail.mcgill.ca

The chemistry involving mercury and organic matter entails interfacial and heterogeneous interactions. In heterogeneous mixture of dissolved organic matters (DOMs) mercury and its related compounds can form complexes, which can exhibit enhanced photoreduction capability. Furthermore, surface interactions can catalyze these reactions. These effects can influence the overall mercury cycling. Therefore, we took interest in deciphering the chemical identity of mercury compounds and their corresponding reactions in the presence of organic matters from a fundamental perspective. Using traditional spectroscopic methods and a laser based surface selective spectroscopic technique we have studied mercury interactions with model heterogeneous particulate system in the aqueous phase. We will discuss the impact of our results on mercury chemical schemes parameterization in the atmospheric global circulation models.

RS9-P2 — 11:00-12:00 and 17:30-18:30
Authors: LIU, Guangliang1, CHEN, Sen1, NAJA, G. Melodie2, CAI, Yong1, TACHIEV, Georgio1, ROELANT, David1
(1) Florida International University, liug@fiu.edu; (2) Everglades Foundation;

Thiol (-SH) containing substances, including low molecular weight (LMW) amino acids and organic matter (OM), play an important role in regulating the speciation, reactivity, and bioavailability of mercury (Hg) species, including inorganic Hg (iHg) and methylmercury (MeHg), thanks to the strong affinity of thiol towards Hg species. The interactions of mercury species with OM have received much attention and many aspects of mercury-OM interactions (including the complexation of Hg with thiol groups within OM) have been extensively studied, whereas the interactions of Hg with LMW thiol-containing compounds present in the environment are investigated to a much lesser extent. In the presence of different thiol-containing substances, including cysteine, glutathione, and humic acid, the dissolution of mercury sulfide (HgS, red cinnabar), was investigated. The LMW amino acids (both cysteine and glutathione, at the concentration levels from 0.5 to 10 µM) greatly enhanced the dissolution of HgS. Agreeing with previous studies, humic acid can also promote HgS dissolution. Further fractionation of the dissolved Hg species (passing through a 0.22 µm filter) using centrifugal filtration devices suggests that a significant fraction (about 40%) of Hg is in the truly dissolved form (referred to here as Hg passing through a filter with 3 kDa molecular weight cut off, MWCO) in the presence of cysteine or glutathione, probably due to the low molecular weight of these amino acids. In the presence of humic acid, the truly dissolved fraction of Hg in the solution is rather small (less than 5%). The disparity in the phase association and subsequent reactivity of the dissolved Hg species between LMW thiols and OM enhanced dissolution warrants further studies to investigate the effect of HgS dissolution on the transport and transformation of Hg.

RS9-P3 — 11:00-12:00 and 17:30-18:30
Authors: OSWALD, Claire J1, BRANFIREUN, Brian A2, HEYES, Andrew3
(1) University of Toronto, claire.oswald@utoronto.ca; (2) University of Western Ontario; (3) Chesapeake Biological Lab - University of Maryland.

Upland soil landscapes are a significant reservoir of atmospherically-derived mercury (Hg) and an important indirect source of Hg to aquatic ecosystems via runoff in many watersheds. Understanding the spatial controls on Hg storage in upland soil landscapes is essential for accurate predictions of the timing and magnitude of Hg fluxes and how they are related to hydrological and biogeochemical processes. The potential controls on Hg spatial distribution fall into three categories: controls related to deposition; controls related to the accumulation of soil organic carbon (SOC); and controls related to the movement of dissolved organic carbon (DOC) in runoff. In this study, we examined the spatial distributions of historically-deposited ambient Hg and an experimentally-applied Hg isotope (spike Hg), and the stoichiometry of the Hg-SOC relationship in different soil layers across the METAALICUS (Mercury Experiment to Assess Atmospheric Loading in Canada and the U.S.) catchment in northwestern Ontario. Specifically, we tested the hypotheses that canopy density and species type affect the mass of Hg in the soil profile and that there is a downslope increase in SOC and Hg in the soil landscape related to drainage conditions.

Our results showed significantly higher amounts of both ambient and spike Hg under old growth coniferous canopy than under deciduous canopy; however, correlations between the areal mass of ambient and spike Hg and leaf area index, an index of canopy density, were weak and not significant. The spatial distribution of spike Hg in all soil layers was best explained by the spatial patterns of cumulative spike Hg aerial loading. We found weak positive correlations between ambient and spike Hg mass in the well-humified organic soil layer and the topographic wetness index, which is higher in poorly-drained areas than well-drained areas. However, despite strong positive correlations between both ambient and spike Hg and SOC, we found no evidence of a downslope accumulation of SOC, ambient Hg, or spike Hg. Our results suggest that the dominant spatial controls on the Hg distribution in the soil landscape are deposition-related, and more specifically, related to tree species type and the ability of species to accumulate Hg from atmospheric deposition. Our results also suggest that in a topographically complex landscape where soil cover is discontinuous and hydrologic connectivity is limited, vertical fluxes of mercury into the soil profile dominate over horizontal fluxes that translocate mercury downslope.

RS9-P4 — 11:00-12:00 and 17:30-18:30
Authors: FAGANELI, Jadran1, KORON, Neza2, FALNOGA, Ingrid3, SLEJKOVEC, Zdenka3, MAZEJ, Darja3, KOVAC, Nives4
(1) Marine Biological Station, jadran.faganeli@gmail.com; (2) Marine Biological Station ; (3) Jozef Stefan Institute; (4) Marine Biological Stataion.

Marine macroaggregates (macrogels), produced episodically in the northern Adriatic by agglomeration of dissolved organic macromolecules - mostly heteropolysaccharides - of prevalently phytoplankton (diatom) origin, offer a rare opportunity to study interactions between marine macromolecules and metals, i.e. Hg, Cd, Zn, Mn, Cr, Fe, Al, Ni, Co, V, Ba, U, Al, and metalloids, i.e. Se an As. Filtration and centrifugation were used to discriminate between macrogel matrix and interstitial water colloids. The interstitial water fraction was incubated during four weeks to study the macroaggregate degradation. The degradation products were ultrafiltered through membranes with a nominal pore size of 30, 10 and 5 kDa cutoff (MWCO) sequentially in a “cascade fashion”. Ultrafiltrates were characterized using FTIR, organic C (Corg), total N (Ntot) and carbohydrate contents, and CV AFS for Hg and ICP-MS for other metal (Me) and metalloid contents. The highest carbohydrate content and lower C/N ratio were associated with higher molecular weight (MW) fraction (>30 kDa) suggesting that glycoproteins and aminopolysaccharides could be important constituents of this fraction. Conversely, the lowest MW fraction (5-10 kDa) contains lower carbohydrate concentration and higher C/N ratio. Higher Me concentrations and lower Me/Corg. ratios, the latter providing data on Me bonded onto organic matter, in matrix compared to interstitital water colloids suggest that Me in matrix can be also bonded onto metal-containing particles. The studied metals and metalloids, except Se, in colloidal fraction were prevalently bonded onto large macromolecules (>30 kDa). During the fast degradation of all except the lowest MW fraction, the Me associated with high MW (>30 kDa) fraction are markedly preserved while Me in lower MW fractions, mostly composed of polysaccharides, are prevalently released into solution. Consequently, the Me/Corg. ratios increased due to the loss of Corg. and relative increase of metal-binding sites. Despite our metal series are generally in a good agreement with Irving-Williams series slight differences can be explained by variable background (“dissolved”) concentrations of studied metals and metalloids.

RS9-P5 — 11:00-12:00 and 17:30-18:30
Author: NAVRATIL, Tomas1
(1) Inst. of Geology AS CR, v.v.i., navratilt@gli.cas.cz

The mining and combustion of high-sulfur coal in the Black Triangle region of the northwestern Czech Republic has left a well-documented legacy of sulfate pollution, but the fate of associated mercury (Hg) has received little attention. We investigated the export of Hg at Lesní Potok (LP), a 75-ha forested catchment about 60km S of the Black Triangle and 30km east of Prague. The LP catchment is 99% forested with 50% deciduous and 50% coniferous cover. The soils developed on the granite are relatively well-drained Cambisols. Precipitation averaged 680mm during the last 10 years and runoff averaged 90mm thus evapotranspiration was high at 87%. LP stream water is dominated by SO42- and Ca2+ with mean concentrations of 59.8 and 12.4 mg.L-1, respectively in water year 2009. The pH ranged from 5.2 to 6.0.

Mean concentration of Hg in soil at LP catchment was somewhat elevated at 330 µg.kg-1 in organic horizons and near background at 54 µg.kg-1 in mineral horizons. Filtered total Hg concentration in LP stream water in water year 2009 was very high, averaging 17.1 ng.L-1 (range 3.2 to 27.3 ng.L-1) DOC averaged 10.3 mg.L-1. Filtered total Hg positively correlated with DOC concentrations (r=0.57; n=29, p<0.001). The usual sampling protocol at LP catchment includes 12 samples per year. Since the spring snowmelt period is usually the most important period of water year in central European forested ecosystems, we increased sampling frequency and took 20 extra samples in 45 days (27 February-13 April 2009). Mean filtered Hg concentrations of 18.0 ng.L-1 in the stream water during snowmelt period were slightly greater than mean annual values but mean DOC concentration did not change significantly (10.6 mg.L-1). This shift in filtered Hg concentrations weakened correlation with DOC (r=0.47; n=22, p<0.01). Annual output of filtered Hg from LP catchment was estimated at 0.87 µg.m-2, with more than 70% of annual Hg output flux occurring during spring snowmelt period. Without the frequent sampling during snowmelt, annual Hg output flux would have been underestimated at 0.79 µg.m-2. The mean ratio of Hg(ng)/DOC(mg) during the snowmelt was 1.70, while during the rest of the year it ranged from 0.45 to 2.03. The shifts in ratio suggest a possible various sources of DOC and Hg. Our next phase of research will address the cause of this very high Hg/DOC ratio and its seasonal shifts.

RS9-P6 — 11:00-12:00 and 17:30-18:30
Authors: HEYES, Andrew 1, GHORPADE, Sarah2, GILMOUR, Cynthia C3
(1)University of Maryland Center for Environmental Science, heyes@umces.edu; (2) University of Toronto; (3) Smithsonian Environmental Research Center.

Mercury (Hg) in watersheds is strongly associated with soil organic matter. As a result, Hg retention is associated with soil formation and mobility with the release of dissolved and particulate organic matter. In Boreal systems, dissolved organic matter (DOM) is the main vector of Hg transport. As components of DOM are differentially reactive, the nature of DOM ligands carrying Hg will influence the fate of Hg within various watershed compartments.

Using the METAALICUS watershed, we examined the solid phase partitioning of newly-deposited Hg (applied as an enriched Hg isotope) and the existing ambient pool of Hg in different soil horizons using a sequential extraction approach as a proxy for Hg-DOM solubility and potential mobility. The applied Hg isotope is largely constrained to the litter and organic soil horizons, but is most easily extracted from the litter layer. The organic soil horizon contains the highest concentration of ambient Hg and the largest per gram amount of easily extractable ambient Hg.We characterized the DOM by molecular weight fractionation and lignin phenol composition, in water collected from different hydrological settings; soil porewater, flow path convergence zones and streams. While low molecular weight (LMW) DOM compounds were more mobile, high molecular weight (HMW) fractions contained both more ambient Hg and enriched Hg isotope per unit carbon. Source material composition, as indicated by lignin phenols, did not vary widely and was not well correlated with Hg.

In summary we found that potential Hg solubility, as determined by sequential extractions, is greatest in the soil organic horizon. The enriched Hg isotope, which has been resident on average for only 5 years, has not yet penetrated into the soil profile to show the same distribution as ambient Hg and thus does not have the same mobility potential. While the Hg isotope is measurable in the soil porewater, little is currently being transferred through surface and subsurface flow paths. In the soil porewater, both ambient Hg and enriched Hg isotope are preferentially associated with high molecular weight DOM but it is low molecular weight DOM that is more mobile. The molecular weight bias of Hg within DOM may explain the rapid loss of both ambient Hg enriched Hg isotope before entering major flow paths or streams. Loss, either by retention or degradation of HMW DOM thus retards Hg export from watershed.

RS9-P7 — 11:00-12:00 and 17:30-18:30
Authors: EKLÖF, Karin1, FÖLSTER, Jens1, SONESTEN, Lars1, BISHOP, Kevin1
(1) Swedish University of Agricultural Sciences, karin.eklof@slu.se

Since the year 2000, 19 Swedish watercourses have been sampled every second month for total mercury (THg) and general water chemistry. The sites were selected for monitoring focus on long term trends in catchments not effected by intensive forestry, local industrial discharge or urban areas.

Approximately 30 potential explanatory variables including water chemistry parameters, hydrology and catchment properties were used in partial least squares (PLS) regressions to evaluate factors affecting the THg export. A humic/organic matter (OM) gradient formed the first component that explains most of the variation in the THg data. Variables measuring the OM content, absorbance at 420 nm (Abs420) and total organic carbon (TOC), explained 66 % and 61 % of the variance respectively. Discharge, soil Hg concentrations and forest cover were also positively correlated with both THg and TOC.

The fact that THg and OM are strongly correlated even when data from all streams in different parts of the country, with different Hg deposition levels and climate were used in a single regression model indicates that OM is the dominant factor controlling the Hg export. When normalizing the measurements of THg with TOC, particle concentrations and average discharge in the streams were related to slightly higher THg/TOC ratios, whereas percentage of mires and electrical conductivity were associated with slightly lower ratios.

Despite a general increase in TOC concentrations during the study period, there was no general increase in THg. Consequently, there were decreasing trends in the THg/TOC ratio at almost all sites. The Abs420 did not increase at the same rate as TOC and there were no significant decreasing trends for the THg/Abs420 ratios. Due to the difference in the organic carbon characteristics detected as Abs420 and TOC, the comparison between the trends for these two variables is of high interest. The fact that THg and Abs420 has not increased at the same rate as TOC during the last decade indicates that fractions of OM absorbing at 420 nm are more important for Hg mobilization than other OM-fractions.

RS9-P8 — 11:00-12:00 and 17:30-18:30
Author: GU, Baohua1
(1) Oak Ridge National Laboratory, gub1@ornl.gov

Naturally dissolved organic matter (DOM) is ubiquitous in water and sediment, affecting the redox transformation and speciation of metal contaminants such as mercury (Hg) in the environments. Depending on the redox state of DOM and the environmental condition, DOM can act either as an electron acceptor or an electron donor. Reduced DOM can transfer electrons to oxidized Hg(II) species to form elemental Hg(0), whereas oxidized DOM or DOM subunits may lead to the oxidation of Hg(0) and subsequently complex and stabilize Hg in the environment. In this study we investigate the role of DOM on the redox transformation and complexation of Hg under dark, anoxic conditions. We prepared reduced DOM isolates (both aquatic and terrestrial origins) by abiotic reduction using Pd-coated alumina as a catalyst and H2 as an electron donor. We show that all DOM isolates were capable of rapidly reducing Hg(II) to elemental Hg(0) at relatively low DOM/Hg(II) ratios. However as DOM/Hg(II) ratios was increased, reduction of Hg(II) appeared to be inhibited as demonstrated by decreased production of purgeable Hg(0). This observation is consistent with the self competition of reduction and complexation of Hg by reactive functional groups in DOM (containing both reducing and complexing moieties). The reducing capacity varied among DOM isolates and depended on the Hg(II)/DOM ratio, with the highest value of about 8 mmol Hg/g C. The reduction of Hg(II) is also characterized by an initial rapid decrease in Hg(II) concentrations within the first 15 minutes followed by a slow rise of Hg(II) in 24 to 48 hours, suggesting that reduction is faster than the complexation. Our results indicate that DOM plays a dual functional role in the reduction and complexation of Hg, and the relevant reaction rates are likely control the redox cycle of Hg and its speciation in the anoxic environments.

RS9-P9 — 11:00-12:00 and 17:30-18:30
Authors: BJÖRN, Erik1, POPP, Maximilian2, SKYLLBERG, Ulf3
(1) Umeå University, erik.bjorn@chem.umu.se; (2) AQura GmbH; (3) Swedish University of Agricultural Sciences.

The importance of low molecular mass (LMM) Hg-thiol complexes in Hg biogeochemical processes has recently been highlighted, e.g. by demonstrating enhanced Hg methylation in bacteria cultures when adding cysteine.1 The lack of suitable analytical methodologies to quantify specific LMM Hg-thiol complexes is however hampering, both to establish the presence of such species in various types of natural waters as well as to conduct fundamental studies of their stability and interactions in controlled model systems.

An LC-ICPMS method was developed for the determination of HgII complexes with cysteine, gluthatione, mercaptoacetic acid and 3-mercaptoproprionic acid. Dissolved HgCl2 was incubated with respective thiol under N2-atmosphere and the conditions for synthesis and storage of Hg-thiol complexes were optimized. The molecular structures of synthesized complexes were verified via molecular ion mass determination and isotopic pattern recognition by ESIMS analysis. For the LC-ICPMS method, LC separation modes based on reversed-phase, hydrophilic interaction and ion-exchange were evaluated. The final optimized method was based on reversed-phase LC-ICPMS using a water/acetonitrile step-gradient and species-unspecific isotope dilution analysis to correct for changes in analyte sensitivity during the gradient elution. The separation of the four model complexes and unreacted HgCl2 was accomplished within 15 minutes. The overall yield, including synthesis and column recovery, ranged from 100% for Hg-cysteine to 50% for Hg-mercaptoacetic acid. Detection limits were in the low to sub ng/ml range, making the method useful for a variety of experiments with model systems, but would require the development of a pre-concentration step before expected natural concentrations of the Hg-thiol complexes can be determined. The method was applied to determine the stability of Hg-cysteine and Hg-gluthatione complexes in solution under N2-atmosphere in the presence of inorganic sulfide.

1. J.K. Shaefer, F.M.M. Morel, Nat. Geosci., 2009, 2, 123

RS9-P10 — 11:00-12:00 and 17:30-18:30
Authors: STALLINGS, Kimberly B.1, KERR, Michelle2, NAGY, Kathryn L.1
(1) University of Illinois at Chicago, kstall2@uic.edu; (2) Environmental Protection Agency;

Mercuric sulfide minerals can react with natural organic matter and release Hg(II) to the aqueous environment under oxygenated conditions (1,2). Reducing moieties in the organic matter, such as quinones, have been postulated to have a mechanistic role in the reaction based on correlation of the measured dissolution rates of cinnabar with the aromaticity of the organic matter. We are investigating how the degree of conjugation of quinone compounds affects the extent of reduction of aqueous mercury(II) and dissolution of cinnabar. The hypotheses are that Hg(II) reduction is more rapid and the dissolution of cinnabar is enhanced to a greater extent the less conjugated the quinone structure. Experiments are conducted under controlled reducing conditions. Mercury(II) reduction and cinnabar dissolution are monitored with time by measuring aqueous Hg(II) in the reaction vessel and an inline oxidizing trap solution designed to capture any gaseous Hg(0) released in the reaction. The mass balance of mercury is checked in all experiments. Initial results show that the reduction of aqueous Hg(II) measured in the presence of hydroquinone, juglone, and 2 ,6-anthraquinone disulfonic acid at a 1:50 Hg-to-quinone molar ratio occurs about 50% more effectively by hydroquinone than by juglone. However, anthraquinone disulfonic acid is even more effective than juglone by a factor of 3, indicating the importance of functional groups in addition to degree of conjugation. Enhanced dissolution of cinnabar as a function of the amount of hydroquinone in solution demonstrates that the reaction is surface-controlled. The dissolution reaction appears to stop after a few hours of reaction suggesting that the surface becomes passivated in the experimental systems. The results combined with those of previous studies (1,2) suggest that fluctuations in position of the oxic-anoxic boundary in natural aqueous systems containing dissolved organic matter may result in temporary rapid dissolution of the surfaces of previously-formed mercuric sulfide minerals thereby providing immediate local sources of aqueous Hg(II) regardless of redox conditions.

[1] Ravichandran M., Aiken G.R., Reddy M. M., and Ryan J.N. (1998) Environmental Science and Technology 32, 3305-3311. 69, 1575-1588.
[2] Waples J.S., Nagy K.L, Aiken G.R., and Ryan J.N. (2005) Geochimica et Cosmochimica Acta

RS9-P11 — 11:00-12:00 and 17:30-18:30
Authors: FLECK, Jacob1, ALPERS, Charles N.1, STEPHENSON, Mark2, GILL, Gary3, KRABBENHOFT, David1
(1) U.S. Geological Survey, jafleck@usgs.gov; (2) Moss Landing Marine Laboratories; (3) Pacific National Laboratory;

Wetlands provide many important functions for ecosystems, but they are also known sites for the formation of methylmercury (MeHg), which poses a threat to human and ecosystem health. The formation and concentration of MeHg has been linked to organic matter (OM) concentration and quality in a wide variety of natural systems. Because of this biogeochemical coupling, understanding the linkages across OM and MeHg cycling can provide insights into the processes controlling MeHg in the water column. Over the past five years, the USGS and partners have collected concurrent measurements of aqueous OM, mercury (Hg) and MeHg in a wide range of wetlands within the Sacramento-San Joaquin Delta ecosystem including tidal wetlands, impounded flow-through wetlands, seasonal and permanent ponds, shore-bird ponds and rice fields. Although OM and MeHg were tightly coupled across the range of natural wetlands examined by this study, relationships were widely poor and variable within and across managed wetlands. Causes for the poor relationships were attributed to several mechanisms including 1) the decoupling of OM and MeHg production, 2) differing degradation rates and processes, 3) differences in algal activity and structure, and 4) differences in the diffusive and advective exchange between the sediment and water column. The information gathered through these studies provides insights on the processes controlling MeHg production and transport in wetland systems and suggests simple management actions may be utilized to minimize Hg and MeHg exports to the greater aquatic system.

RS9-P12 — 11:00-12:00 and 17:30-18:30
Authors: MILLER, Carrie1, BROOKS, Scott1, KOCMAN, David1, YIN, Xiangping1, BOGLE, Mary Anna1
(1) Oak Ridge National Laboratory, millercl@ornl.gov

East Fork Poplar Creek (EFPC) in Oak Ridge, TN, USA has elevated mercury concentrations as a result of historical Hg use at the Y-12 National Security Complex (Y-12 NSC). Mercury discharge from the facility has decreased 85% since the 1980’s but Y-12 NSC continues to act as a source of Hg to the system. Contaminated sediments and soils, both inside and outside Y-12 NSC, also act as a source of Hg. The objective of this work was to study the relationship between the quality and quantity of dissolved organic matter and the concentrations of inorganic and methylmercury (MeHg). Filtered (0.2 µm pore size) and unfiltered total Hg (HgT) and MeHg concentrations were measured three times in 2010 and 2011 at 7 sites along a 20-km reach of the creek. In October 2010 the concentration of filtered HgT decreased downstream from 35 to 8.4 ng/L. Unfiltered Hg varied from 125 to 300 ng/L. No decreasing trend was observed downstream likely due to variable suspended solid concentrations throughout the system. In contrast, MeHg increased from 0.10 to 0.25 ng/L and 0.15 to 0.40 ng/L in the filtered and unfiltered fraction, respectively. Filter passing HgT only constituted 7.0 ± 4.5% of the total Hg in the system while MeHg was more abundant in the filter passing fraction (65 ± 13%). The highest filtered (0.88 ng/L) and unfiltered (3.6 ng/L) MeHg concentrations were measured in an ephemeral tributary to the main creek channel draining a portion of the floodplain. Dissolved organic carbon (DOC) concentration increased downstream (1.9-3.5 mg/L) with the highest DOC concentration in the floodplain tributary (4.5 mg/L). Both total and dissolved MeHg were positively correlated with DOC concentration whereas filter passing and HgT were not correlated to DOC. Increasing specific UV absorbance (SUVA254) coupled with decreasing slope ratio (S275-295/S350-400) downstream indicates an increase in molecular weight and aromaticity of the dissolved organic matter (DOM). Filter passing MeHg is positively correlated with SUVA254 and negatively correlated with slope ratio. These results suggest that (i) floodplain soils may serve as a source of higher molecular weight, more aromatic DOM to the creek, (ii) MeHg-particle partitioning varies with DOM quality which may have implications with respect to MeHg mobility and uptake by biota. Additional studies using ultrafiltration are being conducted to examine the distribution of mercury in different size fractions of DOM.

RS9-P13 — 11:00-12:00 and 17:30-18:30
Authors: MOREAU, John1, KRABBENHOFT, David2, AIKEN, George2, RODEN, Eric3
(1) University of Melbourne, jmoreau@unimelb.edu.au; (2) U.S. Geological Survey; (3) .

Sulfate-reducing bacteria (SRB) are a primary producer of methylmercury (MeHg) in the environment. Microbial sulfate reduction readily occurs in sediments of aquatic ecosystems where there is an ample supply of sulfate, and there exists organic rich sediments that support anaerobic conditions necessary for SRB metabolism. In addition, dissolved natural organic matter (DOM) is ubiquitous in aquatic ecosystems and often contains reduced sulfur sites capable of complexing aqueous inorganic mercury (Hg). Given field and lab observations that DOM is generally observed to be positively correlated with MeHg abundance, and neutral HgS(aq) exhibits hydrophobicity, respectively, we hypothesized that presence of a hydrophobic DOM should enhance methylation (i.e., bioavailability of Hg) in the presence of ambient sulfide. To test our hypothesis, we conducted laboratory-based, pure-culture experiments using a known strain of SRB (Desulfobulbus propionicus), stable isotope tracers of Hg, and several different DOM isolates, including two from the Florida Everglades (hydrophobic and transphilic acids); Williams Lake, Minnesota (hydrophobic acid); and, Suwanee River, Georgia (hydrophilic acid).

The experimental results show that although the four different DOM isolates had distinctly different effects on MeHg production rates, and that three of the four DOM isolates produced distinctly greater MeHg production rates than those without DOM (tracer Hg only controls). Among the DOM isolates used, the two hydrophobic acids yielded the greatest MeHg production rates, with rates observed with Williams Lake exceeding the Everglades. The Everglades transphilic acid was the next most effective methylation-facilitating NOM, and the Suwanee River hydrophilic acid had the least positive effect on methylation rates. In fact, the hydrophilic acid tests yielded MeHg rates similar to the Hg only tracer controls. One unplanned observation from these experiments was the revelation that the experimentally applied DOMs contained “natural Hg”, so we were able to examine the relative bioavailability of that Hg versus the tracer Hg. These results revealed that generally greater amounts of natural Hg introduced with the DOM fraction was methylated, revealing that naturally-partitioned mercury was at least as bioavailable as the tracer. Methylation of natural DOM-hosted Hg suggests a commonality with respect to the form of Hg(II) that can be methylated by SRB.

Thursday, 28 July, 2011